NRAO Home > GBT Spectra


glangsto -at-

created on:


An over view of the GBT L-Band (1.07 to 1.87 GHz) gain and Radio Frquency Inteference (RFI) environment is presented. A table of the frequencies of the strongest RFI sources is included.

The uncalibrated spectra below show the differences in L-Band gain with different RF filters. The RF filters are:

  • NOTCH filter, blocking RFI in the frequency range 1.25 to 1.35 GHz.
  • 1.1 to 1.8 GHz band pass filter after second amplifier
  • 1.1 to 1.45 GHz band pass filter after second amplifier
  • 1.3 to 1.45 GHz band pass filter after second amplifier
  • 1.6 to 1.75 GHz band pass filter after second amplifier

The raw spectra are shown below for a number of band pass filter configurations. A number of features are noted:

  • The band is mostly free of strong RFI.
  • A few RFI sources are > 100 times the system temperature.
  • The system has useful gain over entire 1.07 to 1.87 GHz frequency range.
  • Factor of 5 gain variation across the band from 1.1 to 1.8 GHz.
  • Variations in gain with approximately 25 MHz period
  • Small differences between spectra using different converter modules.
  • Fairly constant system temperature over entire band. (See final plot)

The observing setup is summarized at the end of this document. Data are in directory:

Uncalibrated L-band spectra, showing the variation in system gain across the band.

This plot compares two scans using the 1.1 to 1.8 GHz RF filter. The NOTCH filter was used for one scan and not for the other. No other changes were made to the observing system between scans.

Ascii Data (2003_02_15_05:56:50A.6.log)

Comparision of two scans with different RF filters. One scan has the 1.3 - 1.45 GHz filter and the other scan has the 1.6 to 1.75 GHz filter. The NOTCH filter is not included.

Ascii Data (2003_02_15_05:34:46A.4.log)

Comparison of two scans with and without the NOTCH filter. For both of these scans the 1.1 to 1.45 GHz RF filter was selected.

Ascii Data (2003_02_15_05:43:22A.5.log)

RFI Table

The table below summarizes the strongest sources of RFI in the frequency range 1.1 to 1.8 GHz. The table also lists the redshift of neutral hydrogen corresponding to this RFI frequency.
Frequency Frequency Red RFI
Minimum Maximum Shift Source Note
(GHz) (GHz) (z) Name
1.042 1.150 0.278-0.183 Radar Altimeters Minimum 11 PM to 7 AM
1.220 1.234 0.141-0.131 GPS L2 = 1227.60
1.242 1.247 0.126-0.122 GLONASS 1246 + $k\times0.44$ MHz
1.256 1.256 0.116 Radar New Bedford, VA Sweep every 12 sec.
1.292 1.292 0.090 Radar New Bedford, VA Sweep every 12 sec.
1.311 1.375 0.077-0.032 Periodic Narrow lines
1.381 1.381 0.028 GPS house keeping Intermittent
1.417 1.417 0.002
1.424 1.424
1.426 1.426
1.435 1.435
1.447 1.447
1.499 1.501 Laser Range Finder
1.515 1.530
1.548 1.548 INMAR SAT ?
1.552 1.555 INMAR SAT ?
1.557 1.557
1.570 1.580 GPS L1 = 1575.42 MHz
1.602 1.609 GLONASS 1602 + $k\times0.56$ MHz
1.621 1.627 IRIDIUM
1.681 1.686 GEOS SAT ?
1.691 1.691 GEOS SAT ?
1.699 1.710 NOAA WX SAT?
1.724 1.726
Comparision of spectra from signals passing through converter modules 1 and 2 with 5 and 6. These data were obtained in a single scan, and used the 1.1 to 1.8 band pass filter and the NOTCH filter.

Ascii Data (2003_02_15_06:11:05B.11.log)

Comparision of spectra from signals passing through converter modules 3 and 4 with 7 and 8. These data were obtained in the same scan as the plot above.

Ascii Data (2003_02_15_06:11:05D.12.log)

Estimate of the LCP system temperature across the entire 1.1 to 1.8 GHz band, assuming a constant noise diode temperature of 1.7 K.

Ascii Data (2003_02_15_05:55:55A.14.log)

Observing Setup

On 2003 Feburary 15, Don Backer and Glen Langston performed tests of the GBT at L-Band (1.07 to 1.87 GHz) to determine frequency ranges of free of Radio Frequency Interference (RFI). These tests used the L-band receiver with center frequency of 1470 MHz and the GBT spectrometer as the detector. The spectra were obtained in dual circular polarization mode.

The observations presented here were made with the GBT in snow dump position. The L-band receiver was not in the focus position, so the system temperature was high, approximately 60 K.

The spectrometer was set up in the 800 MHz wide, single quadrant mode, with 3 level sampling. For all but the final tests, only the first quadrant was used. The auto-correllations were obtained in dual polarization mode with synchronous detection of the Calibration noise diodes being toggled on and off at a 4 Hz rate.

The IF Rack Optical driver modules 1 and 4 were used for all observations. The wide, 2360 to 3640 MHz, band pass filters in the IF rack were used. Converter modules 1 and 5 were used for all tests.

Band Pass Shape Tests

After switching in the different band pass filters, the IF rack was "balanced". When using the 1.1 to 1.8 GHz band pass filter, the output of the RCP side had to much power. Even with 31 dB of attenuation the optical driver module was showing 3 Volts, about 3 times too much signal.

The NOTCH filter is intended to exclude Radar signals in the band 1.225 to 1.3 GHz.

Converter Module Comparision

For the converter module comparision tests, converter modules 1 to 8 were used. In these tests the spectrometer was set up in the 4 Bank (4 quadrant mode), with each quadrant measuring the auto-correlations for both polarizations. There are only small differences in the Band pass shapes using the different converter modules, analog filters and spectrometer samplers.

System Temperature Test

The final plot above shows the system temperature across the band for the LCP spectra. These data were taken without the notch filter and include the 1.1 to 1.8 GHz RF filter. The data were calibrated assuming a constant noise diode value of 1.7 K, which is the appropriate value for the center of the band, 1.47 GHz.

Home | Contact Us | Directories | Site Map | Help | Search